Defining Soil Materials for 3-D Models of the Near Surface: Preliminary Findings.

摘要

In support of three-dimensional modeling of soils in the near surface, a method was developed to define soil material types quantitatively using statistical comparisons of properties expected to impact sensor performance. To maintain technical continuity with other soils research and ensure relevance to soils engineering, many of the parameters chosen for statistical comparisons included conventional properties familiar to geotechnical researchers. Other, more sensor-specific soil properties, such as effective saturation and thermal conductivity, were used to allow for direct correlation between classified soil materials and sensor response. Initial trials of the method evaluated datasets from four sites in the U.S. and Asia. Early results showed that the number of statistically distinguishable materials tends to mimic the number of soil horizons sampled for a given dataset. Results also revealed a critical need for further research on the interactions of various soil properties and states between and among each other to determine the combinations that have the strongest influence on sensor response patterns. This protocol for delineating near-surface soil materials advances earlier techniques and improves the state-of-knowledge of modeling geologic features in three dimensions for sensor simulation.